ORIGINAL SIN:
A CROSS NATIONAL STUDY OF
THE LEGALITY OF HOMOSEXUAL ACTS
Victor Asal, Associate Professor *
Department of Political Science, Milne Hall
University at Albany: State University of New York
[email protected]
Udi Sommer, Lecturer *
Department of Political Science
Tel Aviv University, Ramat Aviv, Tel Aviv, Israel
+972547137053
[email protected]
Paul G. Harwood
Department of Political Science and Public Administration
University of North Florida
[email protected]
NOTE THIS IS THE COPY SENT TO COMPARATIVE POLITICAL STUDIES AND
NOT THE VERSION THAT WILL BE PUBLISHED AFTER PEER REVIEW
Abstract
This paper is the first to study decriminalization of homosexual acts quantitatively in a cross-national
perspective with a large sample of countries over a period of several decades. Employing path
dependence as its theoretical framework, this work explains how political, economic and legal
institutions at the domestic and the international levels affect the life of individual citizens. The rights
and privileges of individuals, the findings of this study indicate, are determined by a wide array of
variables, including legal origin, economic development, religion, democratization and the position of
the nation in the international community. We use recently released cross-national data concerning
decriminalization of homosexual intercourse, economic conditions and political institutions. A
generalized estimating equation analyzes decriminalization of homosexual acts and a Cox
proportional hazards model examines how long it takes to introduce this legal reform. Lastly, this
study also offers some important lessons about civil rights and liberties more generally.
* The first two authors had an equal share of the work on this paper and their names appear in
alphabetical order.
KEYWORDS: Sodomy Law; Common Law; Path Dependence; Comparative Public Law;
Legal Evolution;
In 1998, while responding to a report on a shooting in a private dwelling, the Houston police
entered the apartment of John Lawrence. Upon entering the residence, the police found Mr.
Lawrence, an adult man, engaging in consensual homosexual activity with another adult male by
the name of Tyron Garner. Lawrence and Garner were arrested and convicted on charges of
violating a legal ban on sodomy in the state of Texas. The State Court of Appeals upheld the
provision, prohibiting two adults of the same sex from engaging in certain sexual acts. The Texas
court used a 1986 decision of the U.S. Supreme Court, Bowers v. Hardwick, as controlling
precedent. However, when the U.S. Supreme Court reviewed Lawrence v. Texas five years later,
the justices handed down a 6-3 decision, not only reversing the decision of the State court, but
also overruling Bowers v. Hardwick. In 2003, the U.S. Supreme Court’s decision in Lawrence
legalized consensual same-sex intercourse in the United States. Yet, prohibitions against sodomy
still exist in ninety-three nations worldwide. In seven of those, individuals may pay with their
lives for engaging in homosexual activity. This paper is the first to quantitatively study the
decriminalization of homosexual activity cross-nationally with a large sample of countries over a
period of several decades (1972-2002).
Our key argument is based on legal path dependence. We contend that nations with legal
systems based on English Common Law inherited a prohibition of homosexual acts, which
influenced the provision of gay rights—in some cases—for centuries to come. For instance,
when it became part of the British Empire in 1840, homosexual intercourse became illegal in
New Zealand. The prohibition against any type of sexual relations between men was expanded
later in the 19th Century, and it was not until 1986, with the amendment of the Crimes Act and
the passage of the Homosexual Law Reform Act, that consensual sex between men over the age
2
of sixteen became legal. Other examples include Australia, India1, and Israel (Joseph 1996;
Sanders 2009). Using data drawn from “State-sponsored Homophobia: A World Survey of Laws
Prohibiting Same Sex Activity between Consenting Adults” (Ottoson 2009), we examine the
questions of why, and when, nations decriminalize homosexual acts.
This paper serves as the first large-scale quantitative cross-national study of the
decriminalization of homosexual acts. While there has been some work on the legality of
homosexual relations in particular countries (Healey 2002; Schmid 2000), particularly within the
states of the United States (Eskridge 2008; Kane 2003; 2007; Pinello 2003; Robertson 2006), and
some comparisons between countries in the West (Ben-Asher 1989; Wald and Green 1997;
Hensle 2009), there has been very little cross-national inquiry of the legality of homosexual acts
outside of the West (Sanders 2009). There also exists a paucity of time-series cross-national
quantitative analysis beyond the United States2.
The findings in this paper illustrate how political, economic, and legal institutions influence
the lives of individual citizens. Our study indicates that the legality of homosexual acts is
determined by a wide array of variables, including the type of legal system3, economic
development, religion, democratic conditions, and globalization. While the focus here is the
rights of a minority defined by sexual orientation, this study also offers some important lessons
about civil rights more generally (Haider-Markel and Meier 1996; Nownes 2004, inter alia).
This paper is important theoretically, but also significant from a policy perspective. In a
world where more than ninety countries criminalize the consensual sexual activities of same-sex
1
A Delhi High Court ruled in July 2009 in Naz Foundation v. Government of NCT Delhi that Section 377
of the Indian penal code was unconstitutional and thus effectively decriminalized homosexual
relations between consenting adults.
2
Much of the literature that does exist takes a particularly legal perspective and does not attempt to
address possible larger causal mechanisms.
3
We use Type of Legal System and Legal Origin interchangeably (La Porta et al. 2008).
3
adults, and where such activity can, in some instances, lead to the death penalty (Ottoson 2009),
a better understanding of the causes of such criminalization and the factors that lead to the
decriminalization of homosexual activity is important.
What is more, the legality of homosexual acts is only one piece of the puzzle. It is
important to recognize at the outset that countries where homosexual acts are legal are not
necessarily places where homosexuals are treated equally. For example, less than thirty countries
offer some kind of civil union for same-sex couples (Ottoson 2009). Legality, hence, is not the
ultimate measure of discrimination against sexual minorities. With that in mind, an examination
of the legality of homosexual activities is key to developing a better understanding of the rights
of homosexuals cross-nationally.
LEGAL PATH DEPENDENCE
Path dependence refers to the “dynamics of self-reinforcing (…) processes in a political system”
(Pierson and Skocpol 2002, 699). This type of process is widespread in politics (Stinchcombe
1965; Pierson 2000). Once certain institutional rules are in place, alternatives forgone in earlier
stages of the historical process cease to be available (Shepsle 1986). Path dependence appears in
the evolution of laws and legal systems (North 1990) but applies also to a variety of other
political phenomena (Collier and Collier 1991; Kurth 1979; Ertman 1997; Huber and Stephens
2001; Skowronek 1993; Shefter 1977; Skocpol 1979).
We argue that legal path dependence is a key to understanding decriminalization of
homosexual intercourse cross-nationally. The original sin related to homosexual acts (to the
extent that there is one) is related to the legal system put in place by the state, rather than to the
behavior of individuals. The “Original Sin” linked to the criminalization of gay sexual activity is
that of the export of the Common Law system that criminalized buggery in Great Britain in
1533. Common law adopted by other nations (or alternatively imposed on them) in conjunction
4
with subsequent judicial decisions and statutes passed over the centuries, led to criminalization
of homosexual acts (Sanders 2009). Conversely, the adoption and “export” of other legal
systems, particularly French Civil Law, did not have this effect, as sodomy was decriminalized
immediately following the French Revolution. As a result, countries that were less influenced by
British law were much less likely to criminalize same-sex relations.
Two major self-enforcing mechanisms help entrench one legal system over alternative
ones. The first is large setup (or fixed) costs. Setting up a new system of law entails costs, which
short of a major event, such as a revolution or occupation by a foreign nation, are prohibitively
high. Theses costs, therefore, considerably diminish the chances of implementing a new system
of law. The second mechanism involves adaptive expectations. Once a certain legal system is in
place (e.g., after a territory is occupied and colonized), the belief that this system will persist is
enhanced (North 1990). Further, the longer the system persists, the stronger becomes the belief.
The results of these self-reinforcing mechanisms are characterized by multiple equilibria,
possible inefficiencies, lock-in, and path dependence. Multiple equilibria connote that more than
one legal system is possible and the final outcome uncertain. Hence, a potentially better solution
(e.g., a superior legal system) may not win adherence because of the initial advantage of the
existing system. What is more, once established, an existing legal system is “locked-in” and thus
difficult to change. Consequently, increasing returns and significant transaction costs due to
imperfect markets shape institutional change. Once an equilibrium solution is locked-in, change
happens in a particular path, determined by legal and political institutions. This path dependence
is a way “to narrow conceptually the choice and link decision making through time” (North, p.
98).
Path dependence applies to the evolution of law in a variety of ways and Common Law is
a prominent example of such institutional development (North 1990). For instance, one reason
why legal systems are hard to change is the interdependence between legal education, case law,
5
judicial selection, and the existence of juries and other major institutions. Thus, it would be
difficult to change from a system based primarily on codes to one based on cases. Doing so
would require fundamental changes such as reforming legal education and the retraining of
magistrates, judges and justices. Along the same lines, the adoption of a jury system in systems
with no juries would require among other changes, a distinction between law and fact, trials
which occur on consecutive days, and judges and lawyers trained to deal with juries. Granted,
certain laws are not locked-in. Speed limits, for example, are not entrenched in the system. In
most countries, those limits may be changed easily. However, by its nature, the legal framework
of Common Law has the capacity to lock-in laws, particularly if different players have vested
interests in those laws. Those vested interests may be economic, moral, religious, or
sociopolitical in nature.
The legal framework in Common Law consists of constitutional provisions, acts of the
legislature, and judicial decisions. It is a precedent-based system in which past decisions are
integrated into the law; as such, each case sets a legal precedent. Therefore, the law changes
marginally as new cases are decided and a new legal precedent is set, thus becoming part of the
legal structure. Judicial decisions “reflect the subjective processing of information in the context
of the historical construction of the legal framework” (North, p. 97). The Northwest Ordinance,
for instance, passed in 1787 by the Continental Congress, illustrates path dependence and the
historically derived continuity it implies. The Ordinance stemmed from preexisting English and
colonial provisions and concerned government and settlement of the West and its integration
into the new nation. Due to the vested economic and political interests of various parties, for
decades to come, the political and economic structure in the territories was derived from the
framework dictated by the Ordinance. Along the same lines, for political, moral, religious, and, in
some cases, economic reasons, the legal status of gay marriage or abortion law in the United
States is difficult to change.
6
The legality of homosexual acts in Common Law systems, we argue, was heavily
influenced by path dependence. More specifically, the effect of the English Buggery Act of 1533
on the legal status of sodomy was somewhat similar to the effect of The Northwest Ordinance
on American expansion to the West. Legal prohibitions against sodomy were entrenched in
English law (Rayside 1992). While certain laws prohibiting homosexual acts in the military
appeared as early as the eleventh century, it was Henry VIII of England who introduced such
legislation into English criminal law in the form of the Buggery Act of 1533. The Buggery Act
made buggery (referring to an act of sodomy) punishable by death (Dundes 2002). It was not
until 1861 that the capital punishment prescribed by the law was removed4. The Sexual Offences
Act of 1967 provided limited decriminalization of homosexual acts. The Act implemented some
of the recommendations in the Wolfenden Report5 published in 1957. Thus, it took ten years of
campaigning for only a subset of the recommendations in the Report to be etched into law. With
processes such as colonization, these prohibitions found their way into other countries with
Common Law systems and, by ways of path dependence, influenced how laws concerning
homosexuality and homosexual acts developed in those jurisdictions.
Conversely, in other colonial empires, such as France, homosexual acts have been legal
since 1791. The criminal code drafted by the National Constituent Assembly after the French
Revolution rejected the definition of crimes based on the proscriptions of the Christian religion.
Homosexual acts were thus not mentioned in the new penal code. The Napoleonic Code of
1804 and its subsequent Penal Code of 1810 did not undo the decriminalization of homosexual
intercourse.
4
Section 61 of the Offences against the Person Act 1861
5
The Report of the Departmental Committee on Homosexual Offences and Prostitution (better
known as the Wolfenden report) was commissioned by the British government after a series of
prominent people were convicted for homosexual behavior. It recommended that homosexual sex
no longer be a criminal offence (Devlin 1965).
7
French civil law was introduced in many countries under French occupation during the
Napoleonic Wars. Due to path dependence, this has had extensive influence in Europe and
because of colonization, beyond the continent as well (e.g., in many Latin American countries,
many of which adopted it voluntarily after the Spanish occupation was over). Notably, path
dependence applies to the case of French Code just as well. As provisions prohibiting
homosexual acts were absent from the Napoleonic Code of 1804 and the Penal Code of 1810,
there was no “original sin” entrenched in the legal system in nations that adopted Civil Law.
Hence, law developed in a different path.
Whereas Great Britain was exporting a legal system that outlawed sodomy, French code
(which in various forms was imposed by France, Spain, and the Netherlands on their colonies, or
adopted voluntarily by many other states) did not contain such a provision. For centuries to
come, its influence on the way legal and political institutions dealt with the issue of homosexual
acts was, thus, fundamentally different. As for systems with origins in Communist or Socialist
law, many of these states have gone through a process of democratization in the 1990s. As a part
of this process, those states adopted a civil law system. While some of those states had had antisodomy laws in place, the process of democratization in many cases was followed by
decriminalization of gay sexual activity.
A Common Law system is not the only legal variable to increase the likelihood of antisodomy laws, as some religions proscribe homosexual acts. Establishment of a state religion, or
even more so, a legal code that stems from religious principles, increases the likelihood that legal
proscriptions against homosexual acts will be codified. For example, on May 11, 2001, the
Egyptian police arrested fifty-two men. The group, who later became known as Cairo 52, was
aboard the Queen Boat, a floating gay nightclub. While all fifty-two pleaded innocent, they stood
trial, some of them more than once. They were found guilty of charges ranging from “habitual
debauchery” to “contempt for religion.” Even in the face of international criticism, the courts
8
carried on with the trials, in a nation where Islam was the state religion and homosexual
intercourse was illegal.
Based on the principles of the Qur’an, the central text of Islam, and Hadith, which are oral
traditions determining the Muslim way of life, Islam proscribes homosexual acts. Like in versions
of Judaism and Christianity, from its beginning as a religion, Islam rejected homosexual
intercourse. Some nations with a Muslim majority but which are relatively secular in nature
and/or multi-religious (e.g., Indonesia) do not treat homosexual activity as a crime; as such, there
are no prohibitions against it in their legal codes. In some cases (e.g., Turkey), tolerance to
homosexual sexual activity has been entrenched in the system for years (Murray and Roscoe
1997). Yet, nations in which Islam is the state religion (e.g., Egypt), and in Islamic states where
Sharia law and the Qur’an are the primary sources for legislation, there typically exists a codified
prohibition of homosexual activity. Furthermore, in several of those states (e.g., Saudi Arabia,
Somalia, and Mauritania), death is the prescribed punishment for engaging in such activity.
Taken together, the legal code and state religion create a powerful explanation as to why
certain states are less prone to decriminalize sodomy. Due to path dependence, nations with
Common Law systems operate within a legal framework in which homosexual intercourse is a
criminal act. Likewise, in Islamic law, sodomy is prohibited. Yet, certain variables reduce the
costs associated with changing the legal status quo. As a result, they increase the likelihood that
decriminalization will occur. Our key argument is that the type of legal system (Common Law or
Islamic Law) had an influence on the evolution of gay rights in the different political systems.
However, the effect of this “original sin” was gradually mitigated as nations became more
democratic, provided better representation for minorities, globalized and as their economy
developed.
9
EXPLAINING THE LEGALITY OF HOMOSEXUAL ACTS CROSS-NATIONALLY – LITERATURE REVIEW
AND HYPOTHESES 6
To explain prohibitions against homosexual activity, we propose that there are four primary and
interrelated factors at play: the type of legal system; the democratic conditions and the political
opportunity structure for minority populations; the state of economic development within the
country; and the extent to which a country is globalized socially, economically, and politically.
We hypothesize that these dynamics jointly influence whether a nation has chosen to codify
prohibitions of homosexual activity.
Type of Legal System
Due to path dependence we expect nations with a legal system based on English Common Law
to be more likely to ban homosexual relations. The other nations in our data have Civil Law.
Some of them have had this system for centuries. In addition, many of the republics of the
former Soviet Union changed their originally Communist or Socialist systems to Civil Law upon
democratization in the 1990s. The Common Law dummy variable, thus, equals 1 for nations with a
Common Law system. The value is 0 otherwise.
H1: Ceteris paribus, nations with a Common Law legal system should be more likely to ban homosexual
relations.7
Sanders (2009) stresses that much of the prohibitions against homosexual activity in
Islamic nations were imported from Western common law systems (p. 12-15). We do not
dispute this contention, yet we argue that apart from Common Law, Islam should have an
independent effect.
6
All data except the data on sodomy laws were taken originally from the Quality of Government
Database (Teorell 2008); however, all original data sources are cited as such.
7
We use data from La Porta et al 1999 to test this hypothesis.
10
H2: Islamic states and nations where Islam is the state religion are more likely to prohibit same-sex
relations.8
Akin to Islam, Catholic dogma proscribes homosexual acts as sinful and contrary to
natural law (Dempsey 2008). Catholicism is officially endorsed as a state religion in certain
countries. Yet, unlike Islam, it is not likely to be a major source of legislation given the greater
history of independent government action in Western Europe and Latin America. However, we
predict that the higher the Percentage of Catholics in the Population, the greater the likelihood that the
state will criminalize homosexual activity.9
H3: The more sizeable the percentage of Catholics in a population, the higher the probability that laws proscribing
homosexual behaviour will exist.
Democratic Conditions and Political Opportunity Structure for Minority Populations
Based on the democratic understanding of consensual sexual activities between adults as a
human right (Mertus 2007), we expect the spread of democracy to influence this right (Gurr
2000; Davenport 1999). Along the same lines, there is a great deal of empirical analysis to
support the contention that established democracies are more likely to make greater legal
provisions for historically marginalized populations (Davenport 1999; Gurr 2000; Wilensky
2002). To measure Democratic Conditions, we utilize the POLITY score, which was imputed using
Freedom House data where it was missing. The scale ranges from 0 (least democratic) to 10
(most democratic) (Hadenius and Teorell 2005).
H4: The more democratic the polity, the more likely it is that homosexual activity will be legalized.
An examination of political inclusion and political opportunity structures is essential to
understanding the legal prohibitions of homosexual activity. More specifically, we look to the
inclusion of women in the political process (Grey 2002; Meyer 2003; Swers 2001; 2002; Thomas
8
To test this hypothesis, we use data from the CIA World Factbook (September 2009).
9
We use data from La Porta et al 1999.
11
1991; Vega and Firestone 1995). Women’s inclusion is a good proxy for the political inclusion
of the other – particularly in relation to the issue of gender (Robinson and Spivey 2007; Seidman
2009). As a measure of political inclusion and political opportunity structure, we examine the
Percentage of Women Legislators in national legislative bodies. To measure this we use the InterParliamentary Union Women in National Parliaments data (Inter-Parliamentary Union 2005).
We use the data for the legislative body as a whole in unicameral systems, and where there exists
more than one legislative chamber, we use only the percentage of women in the upper house10.
H5: As the number of women in the legislature increases, the likelihood of legal prohibition of homosexual
activities decreases.
The State of Economic Development
As economic conditions improve, we expect the decriminalization of homosexual activity to be
more likely. Political inclusion and political opportunity structures are related to changes in the
economic composition of a state (Doorenspleet 2004; Huntington 1991; Lipset 1960;
Przeworski, Limongi Neto et al. 1997). To measure Economic Development, GDP per capita in
constant US dollars at base year 2000 was used. Missing data were imputed by using the CIA
World Fact Book and through extrapolation (Gleditsch 2002).
H6: As GDP per capita increases, the likelihood that homosexual activity is criminalized should decrease.
Globalization
Conventional theories of sovereignty have emphasized the nation-state as the guarantor of rights
and liberties (Weber 1994 [1919]; Anderson 1983). Yet, some have argued that globalization
erodes the nation-state and leads to porous national boundaries in terms of legal arrangements,
capital flow, immigration and in a variety of other ways (Spivak and Butler 2007; Grewal 2005).
Globalization reduces transaction costs across a range of human interactions, opening up new
opportunities and exposing social systems to new ways of thinking (Hollingsworth 1998). In
10
Ideally we would use the percentage in the lower house, however, such data are not available.
12
terms of the path dependence of legal change, information about alternative legal arrangements
should be more readily available in a globalized state. Consequently, it becomes easier to identify
alternative legal frameworks and the way civil rights and liberties in general and gay rights in
particular are organized in other jurisdictions. Political entrepreneurs, public opinion (Lax and
Phillips 2009), political organizations, and social movements (Barclay et al. 2009) are then able to
recognize alternative legal arrangements within which to settle gay rights. This in turn would
increase the likelihood of decriminalization of sodomy. In addition, the world culture and
normative diffusion literature argues that the strengthening of links between countries allows for
the diffusion of new human rights and norms of tolerance (Finnemore 1996; Boli and Thomas
1997; 1999; Meyer et al. 1997; Ramirez and McEnaney 1997).
To measure these international connections we use the KOF Index of Globalization
(Dreher 2006; Dreher et al. 2008). The indexes for the globalization variables range from 0-100.
Higher values indicate higher levels of globalization. The overall index of Globalization is the
weighted average of Economic Globalization, Social Globalization and Political Globalization (Teorell et
al. 2009). The measure for economic globalization is defined as the long distance flow of
services, goods, capital, information, and perceptions that accompany market exchanges. This
index not only measures actual flows of trade and investments, but also trade restrictions, such
as tariff rates (Dreher 2006; Dreher et al. 2008). The index of political globalization is measured
by the number of embassies and high commissions in a country, the number of memberships
the country has in international organizations, participation in UN peace-keeping missions, and
the number of international treaties signed since 1945 (Dreher 2006; Dreher et al. 2008). Lastly,
the social globalization measure includes three categories of indicators: personal contacts (e.g.,
telephone traffic and tourism), information flows (e.g., number of internet users), and cultural
proximity (e.g., trade books and number of warehouses of Ikea per capita) (Dreher 2006; Dreher
et al. 2008).
13
H7: Increased levels of globalization (and political globalization in particular) increase the likelihood that
homosexual activity is decriminalized.
Analyzing Legal Change
Apart from explaining why a state’s legal code may prohibit gay sexual activity, we are also
interested in analyzing the likelihood for legal change, that is, when a nation decriminalizes
homosexual activity. Due to path dependence, the duration until decriminalization increases with
Common Law. Furthermore, we expect globalization and economic development to reduce the
costs associated with changing the legal status quo related to sodomy. Unlike in the model for
decriminalization as such, however, in the duration model we do not expect democratic
conditions as such to influence the risk. Instead it is democratization (Vanhanen 2005), a
dynamic process influencing changes in institutional infrastructure as well as alterations in legal
frameworks, that would increase the risk for legal change. Democratization considerably reduces
the costs associated with changing political and legal equilibria, because it opens up the legislative
process at the same time as it challenges previous norms. How far along the nation has gone in
its process of democratization should predict the risk that it repeal a previously existing
prohibition against sodomy. We, thus, control for democratic conditions in this model, but
expect democratization, rather than democracy in the polity, to increase the risk that homosexual
acts are decriminalized.
In the model for legal change, Common Law, globalization, and economic development
are measured as described above. We also control for democratic conditions using the same
variables. To test for democratization, we add Vanhanen’s (2000; 2005) Index of
Democratization, which combines measures of competition and participation as the basic
dimensions of democracy. Competition is measures as the percentage of votes not cast for the
largest party. This measure is multiplied by participation, measured as the percentage of the
14
population who actually voted in the election. The Index varies theoretically from 0 (no
democracy) to 100 and empirically from 0 to 49.
H8: States founded on systems based on Common Law are less likely to decriminalize homosexual behavior.
H9: More globalized states are more likely to overturn legal prohibitions of homosexual activity.
H10: Positive economic development increases the likelihood that a state will overturn legal prohibitions of
homosexual activity.
H11: The process of democratization increases the likelihood that a state will decriminalize homosexual activity.
DATA AND METHODS
Data are taken from the May 2009 report of The International Lesbian, Gay, Bisexual and
Intersex Association on “State-sponsored Homophobia: A World Survey of Laws Prohibiting
Same Sex Activity between Consenting Adults,” and from the University of Gothenburg Quality
of Government Institute’s June 2009 “The Quality of Government Dataset” (Teorell et al.
2009)11. In our combined “Global Sexual Discrimination” dataset, states are categorized into one
of three groups, with a state’s assignment depending on i) whether consensual homosexual
intercourse is legal or illegal within the nation, and ii) if legal, in what year legality was attained.
(Insert Table 1 here)
As Table 1 indicates, our first group—which accounts for the majority of states (93)—details
nations where consensual homosexual intercourse has been outlawed throughout the nation’s
entire existence. Our second group depicts states which have decriminalized homosexual
relations sometime within the last four decades, and the year of legalization; there are 47 such
nations in our data. Our third group lists the 63 nations in which consensual homosexual
intercourse has been legal since at least 1970.
11
All original data sets are also cited.
15
As for legal change, many of the countries decriminalizing sodomy since 1970 did so in the
1990s. As Figure 1 demonstrates, between 1970-2008, the 1990s was the decade with the most
cases of legalization of consensual same-sex intercourse. More than half of the countries in the
database that legalized homosexual acts made this legal reform sometime in that decade. A
considerable number of the countries legalizing homosexual relations during this period were
republics from the former Soviet Union. For instance, in 1998, the year with the greatest number
of cases of legalization, three of the six countries legalizing consensual homosexual intercourse
were Tajikistan, Kyrgyzstan and Kazakhstan.
(Insert Figure 1 here)
We use time-series cross-sectional data, listing all states in the abovementioned dataset for
which data is available for the years from 1972 to 2002. To test Hypotheses 1-7, we employ a
generalized estimating equation (GEE) model (Zorn 2001). A marginal approach, such as the
GEE, is appropriate in this case since we are interested generally in what variables influence
decriminalization of homosexual acts, rather than the propensity to do so in a particular nation,
for which a conditional approach would suffice (p. 475). Since the dependent variable, Legal is
dichotomous (1=homosexual acts are legal; 0= homosexual acts are illegal), and due to the data
structure described above, we employ a GEE model with first-order autoregressive component
and logit as the link function. We use robust standard errors clustered on the country.
Hypotheses 8-11 concern the timing of legal change. In this case, we are interested in estimating
the hazard rate, that is, the “risk” that decriminalization would occur at a certain time, given that
it did not occur previously (Allison 1984). We thus estimate a Cox proportional hazard model to
test the remaining four hypotheses.
16
RESULTS
What factors influence the likelihood that consensual homosexual acts are legal? Table 2 lends
preliminary support to our key hypothesis. As Hypothesis 1 states, the type of legal system is
strongly associated with the way homosexual intercourse is treated within the existing legal
framework. Indeed, almost six out of every ten countries in which homosexual activity is illegal
(the largest group in our dataset) have Common Law systems. Conversely, of the countries in
which anti-sodomy laws were not on the books since at least 1970, over 80% have had civil law
as their legal system.
(Insert Table 2 here)
In addition to the type of legal system, we also identified that the level of globalization
would decrease transaction costs and increase the norms associated with tolerance. According to
Hypothesis 7, globalization would increase the likelihood that homosexual acts would be
decriminalized. Indeed, Figure 2 lends preliminary support to that contention. On the horizontal
axis is the level of globalization, from minimal (left) to maximal (right). The bottom portion of
each column represents the percent of nations within the corresponding level of globalization in
which homosexual acts are legal. As the level of globalization increases, we find that the
likelihood that homosexual acts are decriminalized to increase. Whereas 60% of the nations with
minimum levels of globalization criminalize homosexual activity, only 11% of those with
maximum levels of globalization do so.
(Insert Figure 2 here)
Table 3 presents the results of two generalized estimating equation models. These
multivariate analyses corroborate several of our hypotheses. As suggested in Hypotheses 1 and 2,
having a Common Law legal system significantly decreases the likelihood that the nation has
decriminalized homosexual acts. Likewise, when Islam is the state religion, or when the state is
17
an Islamic state, homosexual acts are very likely to be outlawed. Importantly, even when we
control for the effect of Common Law, Islamic law has an independent effect. This finding
challenges contentions that anti-sodomy laws in the Islamic world are solely a product of
Western influence (Sanders 2009).
Our findings also lend support to our theory concerning globalization. Ceteris paribus,
nations with high levels of globalization, and particularly those with increased levels of political
globalization (Model II), are significantly more likely to decriminalize homosexual activity. What
is more, our theory concerning economic conditions wins support in the data. As Hypothesis 6
states, as gross domestic product (GDP) per capita increases, a nation is less likely to have antisodomy laws on its books.
Lastly, we fail to find support for our “democratic conditions” theory. Surprisingly, neither
the variable measuring political inclusion of women (percent of women representatives in the
legislature) nor the variable gauging the democratic nature of the polity has a statistically
significant effect on the legality of homosexual acts. Hypotheses 4 and 5, therefore, do not find
support in the data.
(Insert Table 3 here)
Table 3 details which factors impact the likelihood of decriminalization of sodomy. In
order to examine the effects of the variables from each of the theories we propose, Figure 3 and
Table 4 include predicted probabilities. These probabilities illustrate the influences of the
variables, which have significant effects in Table 4 (Model I). More specifically, Figure 3
demonstrates the critical influence of Common Law systems. The vertical axis indicates the
predicted probability that homosexual intercourse is decriminalized. The three columns on the
right indicate the predicted probability of decriminalization at different levels of globalization in
nations with a system of Common Law. The corresponding predicted probabilities in countries
where the legal system is not based on Common Law are shown in the columns on the left.
18
As predicted in Hypothesis 1, the probabilities in the columns for nations not based on
Common Law are larger across the board than in Common Law nations. Moreover, the analysis
further substantiates Hypothesis 7 concerning the effect of globalization. As the nation becomes
more globalized, the likelihood that homosexual acts will be decriminalized increases
considerably. For instance, a state with a Common Law system where levels of globalization are
low would have a probability of .05 for decriminalization. Conversely, at high levels of
globalization, a state would have a probability of .36. What is more, higher levels of globalization
in a state that is not based on Common Law system would result in a likelihood of .9. Thus, the
likelihood that consensual same-sex intercourse will be decriminalized increases eighteen fold
when we compare nations with Common Law systems and low levels of globalization to
globalized nations whose legal systems are not based on Common Law.
(Insert Figure 3 here)
In support of Hypotheses 2 and 6, Table 4 illustrates the effects of Islam and GDP per
capita. The numbers in the cells are the probabilities that sodomy was decriminalized. As GDP
increases, so does the likelihood that homosexual acts are decriminalized. Likewise, when a
nation is based on Islamic law or endorses Islam as a state religion, the likelihood of
decriminalization decreases considerably. For example, in a nation with Islamic law where GDP
per capita is lower than $3000, the predicted probability of decriminalization is .07. This
probability is almost ten times greater in rich nations where there is no official endorsement of
Islam.
(Insert Table 4 here)
Analyzing Legal Change
Figure 4 lends preliminary support to Hypothesis 11, which concerns the effects of
democratization on the risk that the legal system would change to allow homosexual relations
between two consenting adults. Repeal of anti-sodomy laws is expected with post-communist
19
democratization. 44% (the largest group) of countries decriminalizing homosexual acts sometime
in the last four decades had communist or socialist law as the origins of their legal systems.
(Insert Figure 4 here)
Table 5 presents the results of a Cox proportional hazards model explaining the risk that a
nation would legalize same-sex intercourse between consenting adults, given that it had not done
so previously. We find statistically significant effects for the type of legal system, political
globalization, economic development and democratization. The effect for democracy in the
polity is in the opposite direction to that hypothesized; however, it is statistically insignificant.
The effects for economic and social globalization, and for number of women representatives in
the nation’s legislature, are marginally significant.
Ceteris paribus, nations with a Common Law system have an instantaneous relative risk of
decriminalization of sodomy, which is 52% less than in nations without a Common Law system.
In addition, an increase of one unit in the political globalization scale, holding all other variables
constant, increases the risk of decriminalization by 3.5%. Moreover, a one unit increase in the
democratization index increases the risk of legal change by 4.5%. Alternative specifications (not
presented here), wherein democratization was excluded, yielded statistically significant results for
democratic conditions. Yet, as the model presented in Table 5 clearly demonstrates, when the
democratization variable is included, there is no effect in terms of democratic conditions.
(Insert Table 5 here)
DISCUSSION AND CONCLUSIONS
In this paper, we developed a theory to explain prohibitions of homosexual acts cross-nationally
and then tested it empirically. Legal path dependence, we argued, explains why a legal ban on
homosexual activity exists and how long it endures. In Common Law systems, legal prohibitions
of homosexual activity were entrenched in the system, with vested interests of various types
making change unlikely. Thus, due to path dependence, where the legal system originates from
20
English Common Law, the likelihood of decriminalization decreases considerably. When legal
institutions prohibiting homosexual activity are in place from the nation’s founding, the costs
associated with changing the legal status quo are high. Likewise, when a state’s legal origin stems
from Islam, a religion that prohibits sexual relations between two consenting male adults, the
likelihood that homosexual activity is legal decreases12. Path dependence also explains why
several factors reduce the costs associated with changing the legal status quo and thus increase
the likelihood of decriminalization. When the nation is globalized, information concerning
alternative legal arrangements is readily available. With economic, social, and political
globalization, it is easier for legal institutions to “travel” between jurisdictions and to be
imported from other nations. Consequently, in a highly globalized nation, it may be easier for
political entrepreneurs and interest groups to utilize public opinion and work successfully
towards legal change (Rimmerman and Wilcox 2007; Canaday 2009). We also find support for
the part of the theory concerning economic development. When GDP per capita is higher,
changing the path determined by pre-existing legal institutions is more feasible. While we find
support for three of the four parts of the theory we proposed to explain the legality of
consensual homosexual intercourse, the most dramatic effects result from combinations of those
predictors. It is the combination of Islamic law and GDP, for instance, that could increase the
likelihood of decriminalization ten fold, and it is the combination of Common Law and
globalization that could multiply this likelihood by as much as eighteen.
As for the risk of legal change, we find support for the hypothesized effects of the legal
system, level of globalization, economic development and democratization. There is less risk of
legal change in a nation with Common Law. Higher levels of political globalization increase the
likelihood that a change in the legal status quo concerning sodomy takes place, given that the
12
That said, in nations with a large Muslim population, which are also secular or multi-religious, or
alternatively where there is a tradition of tolerance, Islam would not have such an effect.
21
prohibition still applies. Higher levels of GDP per capita also increase the risk of legal change.
Notably, when controlling for both democratization and democracy in the polity, we find the
former rather than the latter to have a statistically significant effect.
The lack of effect for democratic conditions in any of the models was puzzling.
Democracy does not have an influence either on whether sodomy was decriminalized or on the
risk that such decriminalization would happen, given that it had not happened in the past. The
“Democratic Conditions” component of our theory, thus, fails to win support. Instead, it is the
process of democratization that is a particularly strong precursor of legal change. An index for
democratization was a strong predictor for an increase in the risk for legal change, and when
included in the Cox proportional hazards model it eliminated the effect of democratic
conditions. The shifts in institutional and political status quos inherent to democratization
increase the likelihood of legal change. This is a major reason why the largest group among
nations experiencing decriminalization in the time period studied are nations with Communist or
Socialist origins to their legal systems. As mentioned, many of those were republics from the
former Soviet Union, which went through a process of democratization during the 1990s. This
process was accompanied in more than one case by repeal of a legal prohibition against
homosexual acts (see also Figure 1).
The theoretical framework employed in this study focuses on legal development as it is
affected by domestic and international institutions, religion, and economic conditions. We do
not intend to discount the significance of public opinion (Brace et al. 2002; Haider-Markel and
Kaufman 2006; Lax and Phillips 2009) or the critical role social movements play for gay rights
(Barclay et al. 2009; Cook 1999; Pinello 2003). In fact, some of the variables we study may be
conducive to shifts in public opinion and may result in higher levels of activity of social
movements. For instance, in a politically globalized nation, the polity is more likely to be
informed of changing global standards with respect to the legality of homosexual activity.
22
Changes in public opinion may follow suit (Frank et al. 2009). Along the same lines, economic
development may result in more resources made available for campaigns (legal or otherwise)
waged by social movements (Courier 2009). Barring issues of data availability, we would include
control variables for public opinion and for the role of social movements in our models.
Our research provides substantial insight into the factors that contribute to the
decriminalization of consensual homosexual relations. In addition, it teaches important lessons
about how legal and political institutions develop and change over time, and about how those
institutions interact with economic conditions and with global trends to influence the very
private lives of individual citizens. The liberty of individuals to engage in the type of sexual
activity they desire, and the right of members of a group defined by sexual orientation not to be
discriminated against, are determined by institutional evolution, which in some cases is centuries
old. What is more, it is global forces and economic trends well beyond the bedrooms of
individual citizens that determine their political rights and liberties.
23
REFERENCES
Allison, Paul D. (1984). Event History Analysis: Regression for Longitudinal Event Data. Newbury Park,
CA: Sage.
Anderson, Benedict. 1983. Imagined Communities. New York: Verso.
Barclay, Scott, Mary Bernstein, and Anna-Maria Marshall. 2009. Queer Mobilizations: LGBT
Activists Confront the Law. New York: New York University Press
Ben-Asher, D. 1989. "Legal Discrimination against Homosexuals in America, and a Comparison
with More Tolerant Societies." New York Law School Journal of Human Rights 7: 157.
Bowers v. Hardwick, 478 U.S. 186
Boli, J. and G. M. Thomas (1997). "World Culture in the World Polity: A Century of
International Non-Governmental Organization." American Sociological Review 62(2):
171-190.
Boli, J. and G. M. Thomas (1999). Constructing World Culture: International Nongovernmental
Organizations Since 1875, Stanford University Press.
Brace, Paul, Kellie Sims-Butler, Kevin Arceneaux, and Martin Johnson. 2002. “Public Opinion in
the American States: New Perspectives Using National Survey Data.” American
Journal of Political Science 46(1)): 173-89.
Butler, Judith and Spivak, Gayatri C. 2007. Who Sings the Nation-State?: Language, Politics, Belonging.
Seagull Books
Canaday, Margot. 2009. The Straight State: Sexuality and Citizenship in Twentieth-Century America.
Princeton: Princeton University Press.
Collier Ruth B. and Collier David. 1991. Shaping the Political Arena: Critical Junctures, The Labor
Movement and Regime Dynamics In Latin America. Princeton NJ: Princeton University
Press
Cook, Timothy E. 1999. “The Empirical Study of Lesbian, Gay, and Bisexual Politics: Assessing
the First Wave of Research.” American Political Science Review 93(3): 679-92.
Davenport, Christian. 1999. “Human Rights and the Democratic Proposition.” Journal of Conflict
Resolution 43 (February): 92–117.
Dempsey, R. 2008. "Theological Opinion The Catholic Church's Teaching About Same-Sex
Marriage." Linacre Quarterly 75(1): 71-82
Devlin, P. 1965. "Morals and the criminal law." The Enforcement of Morals: 10.
Dundes, A. 2002. "Much Ado About “Sweet Bugger All": Getting to the Bottom of a Puzzle in
British Folk Speech." Folklore 113(1): 35-49.
Doorenspleet, Renske. 2004. "The Structural Context of Recent Transitions to Democracy."
European Journal of Political Research 43 (May): 309–35.
Dreher, A. 2006. Does Globalization Affect Growth? Evidence from a New Index of
Globalization. Applied Economics, 38(10): 1091-1110.
Dreher, A., Gaston, N. and Martens, P. 2008. Measuring Globalization Gauging its Consequences. New
York: Springer.
24
Ertman, Thomas. 1997. Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern
Europe. Cambridge: Cambridge University Press
Eskridge, W. 2008. Dishonorable passions: sodomy laws in America, 1861-2003, Viking Press.
Finnemore, M. 1996. "Norms, Culture and World Politics: Insights from Sociology's
Institutionalism." International Organization 50: 325-348.
Frank. David J., Boutcher, Steven A. and Camp Bayliss. 2009. “The Reform of Sodomy Laws
from a World Society Perspective.” In Barclay, Scott, Bernstein, Mary and
Marshall, Anna-Maria. Queer Mobilizations: LGBT Activists Confront the Law. New
York: New York University Press
Gleditsch, K. S. 2002. “Expanded Trade and GDP Data.” Journal of Conflict Resolution, 46: 712724.
Grewal, Inderpal. 2005. Transnational America: Feminisms, Diasporas, Neoliberalisms. Durham, NC:
Duke University Press.
Grey, S. 2002. "Does Size Matter? Critical Mass and New Zealand’s Women MPs." Parliamentary
Affairs 55 (1):19-29.
Gurr, Ted. R. 2000. People vs. States. Washington, DC: United States Institute of Peace.
Haider-Markel, Donald P. and Meier, Kenneth J. 1996. “The Politics of Gay and Lesbian Rights:
Expanding the Scope of the Conflict.” The Journal of Politics 58(2), (May): 332-349
Haider-Markel, Donald P., and Matthew S. Kaufman. 2006. “Public Opinion and Policy Making
in the Culture Wars: Is There a Connection Between Opinion and State Policy on
Gay and Lesbian Issues?,” in Public Opinion in State Politics (ed. Jeffrey E. Cohen).
Stanford, CA: Stanford University
Healey, D. 2002. "Homosexual Existence and Existing Socialism: New Light on the Repression
of Male Homosexuality in Stalin's Russia." GLQ: A Journal of Lesbian and Gay
Studies 8(3): 349-378.
Hensle, E. 2009. "Human Rights and Sexual Orientation: An Expansion of the Equality
Doctrine." Bologna Center Journal of International Affairs 12
Hollingsworth, J. 1998. "New perspectives on the spatial dimensions of economic coordination:
tensions between globalization and social systems of production." Review of
International Political Economy 5(3): 482-507.
Huber, Evelyne and Stephens, John D. 2001. Development and Crisis of the Welfare State: Parties and
Policies in Global Markets. Chicago: University of Chicago Press
Huntington, Samuel. P. 1991. The Third Wave: Democratization in the Late Twentieth Century.
Norman: University of Oklahoma Press.
Inter-Parliamentary Union. 2005. Women in National Parliaments.
http://www.ipu.org/wmne/classif.htm 2004, 20 December.
Joseph, Sherry. 1996 “Gay and Lesbian Movement in India.” Economic and Political Weekly, 31(33):
2228-2233
Kane, M. 2003. "Social Movement Policy Success: Decriminalizing State Sodomy Laws, 1969–
1998." Mobilization: An International Quarterly 8(3): 313-334.
25
Kane, M. 2007. "Timing Matters: Shifts in the Causal Determinants of Sodomy Law
Decriminalization, 1961-1998." Social Problems 54(2): 211-239.
Kurth, James R. 1979. “The Political Consequences of the Product Cycle: Industrial History and
Political Outcomes” International Organization, Vol. 33 (1) (Winter): 1-34
La Porta, R., López-de-Silanes, F., Shleifer, A. and Vishny, R. 1999. “The Quality of
Government”. Journal of Law, Economics and Organization, 15(1): 222-279.
La Porta, R., López-de-Silanes, F., Shleifer, A. and Vishny, R. 2008. "Economic Consequences
of Legal Origin." Journal of Economic Literature, 46(2): 285-332.
Lawrence v. Texas, 539 U.S. 558
Lax, Jeffrey R. and Phillips Justin. 2009.“Gay Rights in the States: Public Opinion and Policy
Responsiveness” American Political Science Review, 103(3)
Lipset, Seymour Martin. 1960. Political Man. New York: Doubleday.
Przeworski, Adam, and Fernando Limongi. 1997. "Modernization: Theories and Facts." World
Politics 49 (January): 155–83.
Mertus, J. 2007. "The Rejection of Human Rights Framings: The Case of LGBT Advocacy in the
US." Human Rights Quarterly 29: 1036-1064.
Meyer, B. 2003. "Much Ado about Nothing? Political Representation Policies and the Influence
of Women Parliamentarians in Germany." Review of Policy Research 20 (3): 401-21.
Meyer, J. W., J. Boli, G.M. Thomas, and F.O. Ramirez. 1997. "World Society and the NationState." American Journal of Sociology 103(1): 144.
Murray, Stephen O. and Roscoe, Will 1997. Islamic Homosexualities: Culture, History and Literature.
New York and London: New York University Press.
Naz Foundation (India) Trust v. Government of NCT Delhi and Others, 2009 (India)
North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge
University Press
Nowness, Anthony J. 2004 “The Population Ecology of Interest Group Formation: Mobilizing
for Gay and Lesbian Rights in the United States, 1950-1998.” British Journal of
Political Science. 34(1): 49-67
Ottoson, D. 2009. “State-sponsored homophobia: a world survey on laws prohibiting same sex
activity between consenting adults.” International Gay and Lesbian Association.
Available at:
http://www.ilga.org/statehomophobia/ILGA_State_Sponsored_Homophobia_2
008.pdf (retrieved 6 August, 2008)
Pedriana, Nicholas. 2009. “Intimate Equality: The Lesbian, Gay, Bisexual, and Transgender
Movement’s Legal Framing of Sodomy Laws in the Lawrence v. Texas Caw.” In
Barclay, Scott, Bernstein, Mary and Marshall, Anna-Maria. Queer Mobilizations:
LGBT Activists Confront the Law. New York: New York University Press
Pierson, Paul. "Path Dependence, Increasing Returns, and the Study of Politics," American
Political Science Review, Vol. 94(2): 251-67.
26
Pierson, Paul and Skocpol, Theda. 2002. “Historical Institutionalism in Contemporary Politics”.
In Katznelson, Ira and Helen V. Milner. Political Science: The State of the Discipline, 3rd
Edition. New York: W.W. Norton & Company.
Pinello, D. 2003. Gay rights and American law, Cambridge University Press.
Rayside, David M. 1992. “Homophobia, Class and Party in England.” Canadian Journal of Political
Science 25(1) (March): 121-149
Ramirez, Francisco O. and Elizabeth H. McEneaney 1997. "From Women's Suffrage to
Reproduction Rights? Cross-National Considerations." International Journal of
Comparative Sociology 38 (June): 6–24.
Ramirez, Francisco O., Yasemin Soysal, and Suzanne Shanahan. 1997. "The Changing Logic of
Political Citizenship: Cross-National Acquisition of Women's Suffrage Rights."
American Sociological Review 62 (October): 735.
Rimmerman, Craig and Clyde Wilcox (eds). 2007. The Politics of Same-Sex Marriage. Chicago: The
University of Chicago Press.
Robertson, S. 2006. "'Boys, of Course, Cannot be Raped': Age, Homosexuality and the
Redefinition of Sexual Violence in New York City, 1880-1955." Gender & History
18(2): 357-379.
Robinson, C. and S. Spivey. 2007. "The Politics of masculinity and the ex-gay movement." Gender
& Society 21(5): 650.
Sanders, D. 2009. "377 and the Unnatural Afterlife of British Colonialism in Asia." Asian Journal
of Comparative Law 4(1): 7.
Schmid, H. 2000. "Decriminalization of Sodomy under South Africa's 1996 Constitution:
Implications for South African and US Law." Cardozo Journal of International and
Comparative Law 8: 163.
Seidman, S. 2009. "Critique of Compulsory Heterosexuality." Sexuality Research and Social Policy:
Journal of NSRC 6(1): 18-28.
Shepsle, Kenneth A. 1986. “Institutional Equilibrium and Equilibrium of Institutions.” In
Political Science: The Science of Politics, Herbert F. Weisberg (Ed.) New York: Agathon
Press
Skocpol, Theda. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia and
China. Cambridge University Press.
Skowronek, S. 1993. The Politics Presidents Make: Leadership from John Adams to George Bush. Harvard
University Press
Stinchcombe, Arthur L. 1965. “Social Structure and Organizations.” In James G. March (Ed.)
Handbook on Organizations. Chicago: Rand McNally, pp. 142-193
Swers, M. L. 2001. "Understanding the Policy Impact of Electing Women: Evidence from
Research on Congress and State Legislatures." PS: Political Science and Politics 34
(2):217-20.
Swers, M. L. 2002. The Difference Women Make: the policy impact of women in Congress: University of
Chicago Press.
Thomas, S. 1991. "The Impact of Women on State Legislative Policies." The Journal of Politics 53
(4):958-76.
27
Teorell, Jan, Nicholas Charron, Marcus Samanni, Sören Holmberg, and Bo Rothstein. 2009. The
Quality of Government Dataset, version 17 June09. University of Gothenburg: The
Quality of Government Institute, http://www.qog.pol.gu.se.
Vanhanen, T. 2000. “A New Dataset for Measuring Democracy, 1810-1998.” Journal of Peace
Research, 37(2): 252-65.
Vanhanen, T. 2005. Measures of Democracy 1810-2004 [computer file]. FSD1289, version 2.0 (200508-17). Tampere: Finnish Social Science Data Archive [distributor].
Vega, A. and J. M. Firestone. 1995. "The Effects of Gender on Congressional Behavior and the
Substantive Representation of Women." Legislative Studies Quarterly 20 (2):213-22.
Weber, Max. 1994 [1919] “The Profession and Vocation of Politics.” In Political Writings. Peter
Lassman and Ronald Speirs (Eds.). pp. 309-369. Cambridge: Cambridge University
Press.
West, Donald J. and Richard Green. 1997. Sociolegal Control of Homosexuality: A Multi-Nation
Comparison. Plenum Press: New York
Wilensky, Harold L. 2002. Rich Democracies: Political Economy, Public Policy, and Performance. Berkeley:
University of California Press.
Zorn, Christopher J. W. (2001). “Generalized Estimating Equation Models for Correlated Data:
A Review with Applications.” American Journal of Political Science 45(2): 470-90
Laws cited:
New Zealand Homosexual Law Reform Act of 1986
British Buggery Act of 1533 (25 Hen. VIII c. 6)
Section 377, Indian Penal Code
British Sexual Offences Act 1967 (1967 c. 60)
28
TABLES
Table 1
The Types of States in the Database (1970-2008)
Country’s
name
Homosexual Intercourse is
Illegal throughout
Change from illegal to legal
status for Homosexual
Intercourse (year of legal
change)
Homosexual Intercourse is
Legal throughout
Afghanistan
Algeria
Andorra
Angola
Antigua and Barbuda
Bangladesh
Barbados
Belarus
Belize
Bhutan
Botswana
Brunei Darussalam
Cameroon
Comoros
Djibouti
Dominica
Egypt
Eritrea
Ethiopia (-1992)
Ethiopia (1993-)
Gambia
Germany, East
Ghana
Grenada
Guinea
Guyana
India
Iran
Iraq
Jamaica
Kenya
Kiribati
Lebanon
Lesotho
Liberia
Libya
Malawi
Malaysia
Maldives
Mauritania
Albania (1995)
Armenia (2003)
Australia (1994)
Austria (1971)
Azerbaijan (2000)
Bahamas (1991)
Bahrain (1976)
Bosnia and Herzegovina
(1998)
Cape Verde (2004)
Chile (1999)
China (1997)
Colombia (1981)
Croatia (1977)
Cuba (1979)
Cyprus (1998)
Ecuador (1997)
Estonia (1992)
Fiji (2005)
Finland (1971)
Georgia (2000)
Guinea-Bissau (1993)
Ireland (1993)
Israel (1988)
Kazakhstan (1998)
Kyrgyzstan (1998)
Latvia (1992)
Liechtenstein (1989)
Lithuania (1993)
Macedonia (1996)
Malta (1973)
Marshall Islands (2005)
Moldova (1995)
Nepal (2007)
New Zealand (1986)
Nicaragua (2008)
Panama (2008)
Portugal (1983)
Romania (1996)
Russian Federation (1993)
Argentina
Belgium
Benin
Bolivia
Brazil
Bulgaria
Burkina Faso
Cambodia
Canada
Central African Republic
Chad
Congo
Congo, Democratic Republic
Costa Rica
Cote d'Ivoire
Czech Republic
Czechoslovakia
Denmark
Dominican Republic
El Salvador
Equatorial Guinea
France
Gabon
Germany, West
Greece
Guatemala
Haiti
Honduras
Hungary
Iceland
Indonesia
Italy
Japan
Jordan
Korea, North
Korea, South
Laos
Luxembourg
Madagascar
Mali
29
Total
Mauritius
Micronesia
Morocco
Mozambique
Myanmar
Namibia
Nauru
Nigeria
Oman
Pakistan
Palau
Papua New Guinea
Qatar
Samoa
Sao Tome and Principe
Saudi Arabia
Senegal
Serbia and Montenegro
Seychelles
Sierra Leone
Singapore
Solomon Islands
Somalia
Sri Lanka
St Kitts and Nevis
St Lucia
St Vincent and the
Grenadines
Sudan
Swaziland
Syria
Tanzania
Tibet
Timor-Leste
Togo
Tonga
Trinidad and Tobago
Tunisia
Turkmenistan
Tuvalu
USSR
Uganda
United Arab Emirates
Uzbekistan
Vanuatu
Vietnam
Vietnam, South
Yemen
Yemen, North
Yemen, South
Zambia
Zanzibar
Zimbabwe
Slovenia (1977)
South Africa (1998)
Spain (1979)
Tajikistan (1998)
Ukraine (1991)
United Kingdom (1982)
United States (2003)
Yugoslavia (1994)
Mexico
Monaco
Mongolia
Netherlands
Niger
Norway
Paraguay
Peru
Philippines
Poland
Rwanda
San Marino
Slovakia
Suriname
Sweden
Switzerland
Taiwan
Thailand
Turkey
Uruguay
Venezuela
Vietnam, North
93
47
63
30
Table 2
Legality of Homosexual Acts by Legal System (1970-2008)
Homosexual
Homosexual
Intercourse is Illegal
Intercourse is Legal
Throughout
Throughout
English Common Law
48
2
Civil Code
28
49
Socialist/Communist laws
6
9
* Numbers in cells indicate number of countries.
31
Table 3
Generalized Estimating Equation Model
Cross-Sectional Time-Series Analyses of Predictors of Legalization of
Homosexual Acts (1972-2002)
Variable
Model I
Model II
(Robust Standard Error)
(Robust Standard Error)
Common Law
-2.78 ***
-2.84 ***
(.44)
(.46)
Islamic State or Islam as
-2.24 *
-2.25 *
State Religion
(1.05)
(1.16)
Catholics (as % of population)
.0059
.006
(.005)
(.005)
.028 ***
n/a
Globalization (General Index
(.006)
Economic Globalization
n/a
.01 *
(.005)
Political Globalization
n/a
.01 ***
(.002)
Social Globalization
n/a
.01 *
(.005)
GDP Per Capita
Percent Women in Legislature
Democracy
Constant
.0000354 *
.000037 *
(.0000157)
(.00002)
-.009
-.01
(.006)
(.007)
.009
.005
(.015)
(.017)
-.87
-.92
(.39)
(.42)
N = 3471
N = 3170
32
Number of Groups = 146
Number of Groups = 133
Observations per Group
Observations per Group
min 4; average 23.8; max 30
min 4; average 23.8; max 30
Wald
Wald
2
Prob >
(7) = 65.38
2
= .000
2
(10) = 69.23
Prob >
2
= .000
Correctly Predicted: 78.8%
Correctly Predicted: 80.1%
Reduction in Error: 42.5%
Reduction in Error: 46.1%
Due to issues of data availability, the following 45 nations were not included in the analyses:
Afghanistan, Andorra, Bhutan, Brunei Darussalam, Comoros, Cuba, Czechoslovakia, Djibouti,
Dominica, Equatorial Guinea, Eritrea, East Germany, West Germany, Guinea, Iraq, Kiribati,
North Korea, Laos, Liberia, Libya, Liechtenstein, Marshall Islands, Micronesia Monaco,
Nauru, Palau, San Marino, Sao Tome and Principe, Serbia and Montenegro, Solomon Islands,
Somalia, St Kitts and Nevis, St Lucia, St Vincent and the Grenadines, Taiwan, Tibet, TimorLeste, Tonga, Turkmenistan, Tuvalu, USSR, Uzbekistan, North and South Vietnam, North
and South Yemen, Yugoslavia, Zanzibar. The truncation is not systematically associated with
any of the predictors as a result --- a third of the nations on this list has or had common law
systems and the averages for GDP, women in legislature, democracy and Catholics in
population for this group of nations and for the sample as a whole are not distinguishable
statistically. Values for the globalization variables are not available for those 45 nations.
Data were not available for some of the countries for some of the years. Data were missing
most typically in the Democracy variable provided by Freedom House.
***p<.001 **p<.01 *p<.05 #p<.1, one-tailed tests where directionality hypothesized.
Robust standard errors are clustered on the nation.
33
Table 4
Predicted Probabilities (for Model I in Table 4)
Effects of Islamic Law and GDP Per Capita
Islamic Law or Islam as State
Not Islamic Law and Islam is
Religion
not State Religion
GDP Per Capita below $3,000
.07
.39
GDP Per Capita between
.13
.59
.19
.67
$3,000-$11,000
GDP Per Capita above
$11,000
* Numbers in cells represent the probability that homosexual intercourse is decriminalized in the state.
34
Table 5
Cox Proportional Hazards Model
Time to Legalization (1972-2002)
Variable
Estimate
(Robust Standard Error)
Common Law
-.73 ***
(.05)
Economic Globalization
.003 #
(.0018)
Political Globalization
.003 **
(.0012)
Social Globalization
.002
(.002)
GDP per capita
.000032 ***
(.0000054)
Women in Legislature
.0051 #
(.0027)
Democracy in Polity
-.007
(.012)
Democratization
.044 ***
(.004)
Number of Observations: 1836
Wald
2
Prob >
(8) = 869.47
2
= .000
Log-Likelihood = -12577.098
The 63 nations where sodomy had been decriminalized prior to the time period studied
were excluded from the analysis by the computer. Furthermore, due to data availability,
mostly related to the democratization and globalization variables, some of the remaining
nations were not included in the duration analysis.
***p<.001 **p<.01 *p<.05 #p<.1, one-tailed tests where directionality hypothesized.
35
FIGURES
Figure 1: Number of Countries Legalizing Homosexual Acts (by year)
36
Figure 2: Percent of Countries with Legal Homosexual Intercourse at Different
Levels of Globalization
Percent of Countries with Legal Homosexuality at Different Levels of Globalization
Homosexuality is Legal Homosexuality is Illegal
100
90
80
70
Percent
60
50
89 %
40
60 %
30
20
40%
41 %
35 %
10
0
Minimal
Low
Average
Levels of Globalization
37
High
Maximal
Figure 3: Predicted Probabilities: Combined Effects of Type of Legal System and
Globalization
Predicted Probabilities: Combined Effects of Type
of Legal System and Globalization
Predicted Probability of Decriminalization of
Homosexuality
Low Levels of Globalization
High Levels of Globalization
Average Levels of Globalization
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Not a Common Law
System
A Common Law System
38
Figure 4: Types of Legal Systems in Countries Overruling a Prohibition on
Homosexual Acts
39
Appendix A
Weibull Regression
Time to Legalization (1972-2002)
Variable
Estimate
(Robust Standard Error)
Common Law
-.66 ***
(.05)
Economic Globalization
.0008
(.002)
Political Globalization
.003 **
(.001)
Social Globalization
.004 #
(.0026)
GDP per capita
.00003 ***
(.000006)
Women in Legislature
.008 **
(.003)
Democracy
-.017
(.012)
Democratization
.04 ***
(.004)
Constant
-32.2 ***
(.85)
Number of Observations: 1836
Wald
2
Prob >
(8) = 609.11
2
= .000
Log-Pseudolikelihood = 766.4
The 63 nations where sodomy had been decriminalized prior to the time period studied were
excluded from the analysis by the computer. Furthermore, due to data availability, mostly related to
the globalization variable, some of the remaining nations were not included in the duration analysis.
***p<.001 **p<.01 *p<.05 #p<.1, one-tailed tests where directionality hypothesized.
40